Self-configured link adaptation using channel quality indicator-modulation and coding scheme mapping with partial feedback for green long-term evolution cellular systems

The continuous growth in wireless communication demands has led to a rapid increase in energy consumption and spectrum utilization. Therefore, a green radio technology is proposed recently to optimize the energy and spectrum resources. One of the main challenges in optimizing the energy and spectrum resources at the cellular downlink transmission is the provisioning of quality-of-service (QoS) when real time traffic is considered. Moreover, improving the energy and spectrum utilization at the downlink while maintaining QoS will require perfect channel state information (CSI) at the transmitting base station (eNodeB). However, reporting the channel status of the whole bandwidth in a multi-carrier communication system requires huge feedback signaling overhead at the cost of uplink performance. Therefore, the aim of this thesis is to optimize the energy and bandwidth resources while maintaining QoS at the downlink when a partial feedback is considered. In this thesis, an analytical model that represents the tradeoff between energy efficiency (EE), spectral efficiency (SE), and transmission delay in both green and conventional Long-Term Evolution (LTE) eNodeB is formulated. Consequently, a modified downlink scheduler based on a Packet Prediction Mechanism (PPM) is conducted at the eNodeB. In the frequency domain of this scheduler, a new mapping between channel quality indicator (CQI) and modulation and coding scheme (MCS) is proposed to find the tradeoff between conflicting criteria. On the user side, a partial channel feedback scheme based on an adaptive CQI threshold is developed. A primary concern of this feedback scheme is to reduce the uplink signaling overhead required by the eNodeB without a substantial loss in downlink throughput and outage capacity. To achieve this objective, an iterative approach based on swarm intelligence is used to find the optimal CQI threshold at which the competing criteria are optimized. Since the developed downlink scheduler and the partial feedback scheme affect the QoS, self-configured versions of both algorithms are developed to provide QoS provisioning. Specifically, the downlink frequency domain scheduler will reconfigure the criteria priorities such that the EE is maximized as long as the QoS is guaranteed.On the other hand, the partial feedback algorithm will search for the threshold value that minimizes the uplink overhead given that the QoS is achieved at the downlink.Otherwise, optimizing QoS parameters will be targeted at the cost of other system parameters. Finally, a self-configured link adaptation with partial feedback (SCLAPF) is developed such that the downlink packet scheduling and the partial feedback are jointly evaluated to further enhance the system performance. The system-level simulation shows that the modified PPM (MPPM) scheduler achieves an improvement in EE of up to 60% compared to the PPM scheduler. On the receiving side, the CQI partial feedback with adaptive threshold reduces the signaling overhead down to 41% compared to the full feedback reporting. Besides, it achieves the most acceptable tradeoff between uplink and downlink performances compared to previous works. More enhancements in downlink and uplink parameters are obtained when the energy efficient MPPM scheduler and the adaptive threshold feedback scheme are jointly implemented. It is shown that the proposed SCLAPF algorithm achieves a high gain in EE of more than 120% compared to that obtained by stand-alone partial feedback performance. Based on simulation results, it is concluded that the proposed energy-efficient scheduling has led the eNodeB to transmit with low power through modified MCS. Besides, it minimizes the degradation in downlink performance which is caused by partial CQI feedback. Therefore, the SCLAPF algorithm, which jointly evaluates the energy-efficient scheduling and partial feedback, gives the best tradeoff between uplink and downlink performances.

Preview PDF FK 2015 106IR.pdf Download (1MB) | Preview

Actions (login required)

View Item